A Comparative Analysis of Intraocular Pressure Measurement across Different Tonometry Methods and Experience Levels
- Author:
Serife Ciloglu HAYAT
1
;
Yusuf Cem YILMAZ
;
Hatice DAVARCI
;
Muhammed Tarik DINCER
;
Sadik Altan OZAL
Author Information
- Publication Type:Original Article
- From:Korean Journal of Ophthalmology 2026;40(3):264-273
- CountryRepublic of Korea
- Language:English
-
Abstract:
Purpose:To evaluate the interobserver reproducibility of Goldmann applanation tonometry (GAT) among ophthalmologists with varying levels of experience and to assess the agreement between noncontact tonometry (NCT) and GAT in measuring intraocular pressure (IOP).
Methods:This cross-sectional observational study, conducted prospectively at a tertiary care hospital, included 210 patients. IOP measurements were performed on the right eye using NCT followed by GAT. Observer 1 was a senior ophthalmologist, while observers 2 and 3 were a second-year and a first-year resident, respectively. The primary objective was to compare GAT-based IOP measurements between observer 1 and the residents; the secondary objective was to evaluate agreement between NCT and GAT.
Results:A statistically significant difference was observed between observers 1 and 2, and between observer 1 and NCT, but not between observers 1 and 3. The mean difference (MD) in IOP between observers 1 and 2 was 0.8 mmHg (95% limits of agreement [LoA], ±5.7 mmHg). Comparisons with observer 3 and NCT yielded MDs of 0.4 mmHg (95% LoA, ±6.5 mmHg) and –0.9 mmHg (95% LoA, ±7.3 mmHg), respectively. A higher percentage of observer 2 readings fell within ±4 mmHg of observer 1 compared to observer 3 and NCT. In the normotensive subgroup, MDs were 0.7, 0.4, and –1.4 mmHg, respectively, with a notably wide LoA for NCT (–8.1 to 5.3 mmHg).
Conclusions:There was notable interobserver variability in GAT-based IOP measurements between residents and the senior ophthalmologist, exceeding the ±4 mmHg clinical acceptability benchmark. This variability suggests that less experienced observers may not yet have fully developed skills, impacting measurement accuracy. Over time, this range may narrow as experience and proficiency improve. Additionally, clinicians should interpret NCT results with caution—especially in postoperative assessments or when modifying treatment. Cross-checking with GAT and incorporating structured training protocols may enhance measurement consistency.
